Transcription Inhibition by Platinum DNA Cross-links in
Live Mammalian Cells
The MIT Faculty has made this article openly available. Please share
how this access benefits you. Your story matters.
Citation
Ang, Wee Han, MyatNoeZin Myint, and Stephen J. Lippard.
“Transcription Inhibition by PlatinumDNA Cross-Links in Live
Mammalian Cells.” Journal of the American Chemical Society
132.21 (2010) : 7429-7435.
As Published
http://dx.doi.org/10.1021/ja101495v
Publisher
American Chemical Society
Version
Author's final manuscript
Accessed
Thu May 26 19:13:42 EDT 2016
Citable Link
http://hdl.handle.net/1721.1/64756
Terms of Use
Article is made available in accordance with the publisher's policy
and may be subject to US copyright law. Please refer to the
publisher's site for terms of use.
Detailed Terms
Transcription Inhibition by Platinum DNA Cross-links in Live Mammalian
Cells
Wee Han Ang, MyatNoeZin Myint, Stephen J. Lippard*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts
02139.
*To whom correspondence should be addressed. E-mail: lippard@mit.edu
ABSTRACT
We have investigated the processing of site-specific Pt-DNA cross-links in live mammalian
cells to enhance our understanding of the mechanism of action of platinum-based anticancer
drugs. The activity of platinum drugs against cancer is mediated by a combination of processes
including cell entry, drug activation, DNA-binding, and transcription inhibition. These drugs
bind nuclear DNA to form Pt-DNA cross-links, which arrest key cellular functions, including
transcription, and trigger a variety of responses, such as repair. Mechanistic investigations into
the processing of specific Pt-DNA cross-links are critical for understanding the effects of platinum DNA damage, but conventional in vitro techniques do not adequately account for the complex and intricate environment within a live cell. With this limitation in mind, we developed a
strategy to study platinum cross-links on plasmid DNAs transfected into live mammalian cells
based on luciferase reporter vectors containing defined platinum-DNA lesions that are either
globally or site-specifically incorporated. Using cells with either competent or deficient nucleotide excision repair systems, we demonstrate that Pt-DNA cross-links impede transcription by
blocking passage of the RNA polymerase complex and that nucleotide excision repair can remove the block and restore transcription. Results are presented for ~3800-base pair plasmids that
2
are either globally platinated or carry a single 1,2-d(GpG) or 1,3-d(GpTpG) intrastrand cross-link
formed by either cis-{Pt(NH3)2}2+ or cis-{Pt(R,R-dach)}2+; where {Pt(NH3)2}2+ is the platinum
unit conveyed by cisplatin and carboplatin and R,R-dach is the oxaliplatin ligand, R,R-1,2diaminocyclohexane.
Keywords: cancer therapy, RNA polymerase II inhibition, nucleotide excision repair, platinum
drugs, xeroderma pigmentosum
3
INTRODUCTION
FDA-approved platinum-based anticancer drugs cisplatin, carboplatin, and oxaliplatin (Figure 1) are among the most effective chemotherapies in clinical application today. They are
widely used to treat a variety of malignancies including colorectal, non-small cell lung, and genitourinary cancers.1-3 The cytotoxic action of these compounds requires a combination of processes including cell entry, drug activation, DNA binding, and cellular responses.4-6 Upon cell entry, the platinum drugs are activated by aquation and bind nuclear DNA, giving rise to a limited
number of adduct types. Cisplatin, in particular, forms predominantly 1,2-d(G*pG*)-Pt (65%)
and 1,2-d(A*pG*)-Pt (25%) intrastrand cross-links, and to a lesser extent 1,3-d(G*pNpG*)-Pt
intrastrand (5-10%) and interstrand (<5%) adducts, where asterisks denote the platinated bases.4,6
Formation of platinated DNA lesions is an important determinant of the anticancer activity of
platinum-based drugs, leading to arrest of key cellular functions and triggering a variety of responses.5 Understanding how the cell processes these Pt-DNA lesions is important for elucidating the mechanism of action of the platinum drugs and forming a basis for the rational design of
improved versions.
One activity that is important for platinum drug-mediated cytotoxicity is the arrest of RNA
synthesis by Pt-DNA cross-links.7 Plasmids expressing the β-galactosidase reporter gene containing platinum-DNA adducts formed by cisplatin inhibit transcription 2-3-fold more effectively
than adducts formed by trans-DDP, the inactive isomer of cisplatin.8 Because platinum drugs
form a variety of DNA adducts, however, it is desirable to develop probes containing a single,
site-specific Pt-DNA cross-link in order to investigate the specific roles of each in mediating biological activity. By introducing site-specifically platinated probes into cell-free extracts derived
from mammalian cells, we previously elucidated the nature of RNA pol II arrest at the site of the
4
platinated DNA lesion.9 Apart from transcription, cisplatin-DNA damage affects nucleotide excision repair (NER), an important pathway for removing the lesions.10,11 NER-defective human
fibroblast cells derived from xeroderma pigmentosum (XP) patients are hypersensitive to platinum drugs.12 These cells contain specific defects in one of seven XP genes, XPA through XPG,
and are generally more sensitive to treatment by platinum drugs compared to normal human fibroblasts. Because the nature of their genetic defect is well characterized, XP cells provide excellent models to study the effects of platinum drugs in an NER-deficient environment.
Until now, mechanistic investigations of platinum-DNA damage processing have largely
been carried out in vitro using site-specifically platinated probes in conjunction with purified enzymes or cell-free extracts.13-19 These systems have been quite informative but they do not adequately replicate the conditions within a live mammalian cell. In order to extend our investigations to this more accurate albeit complex environment, we devised a strategy to study transcription inhibition by Pt-DNA lesions that is based on expression vectors containing single sitespecific platinated DNAs as probes. We sought both to determine the extent to which Pt-DNA
cross-links can inhibit transcription by impeding the passage of RNA polymerase (pol) II in live
cells and the effect of NER in removing the blockage and restoring transcription. To test the
methodology, we first prepared globally platinated transcription probes before moving to the
synthetically more challenging site-specific constructs. The results are described and discussed in
this report.
RESULTS
Methodology. Our methodology is based on the use of platinated DNA plasmids constructed
from a recombinant mammalian expression vector containing a reporter gene under the control
5
of a promoter for subsequent transfection into mammalian cells as functional transcription
probes. Introduction of Pt-DNA cross-links into the plasmids was achieved site-specifically using oligonucleotide insertion. Expression of the reporter gene, unimpeded in the absence of the
lesion, is adversely affected if the lesion is an effective inhibitor of transcription. By determining
transcription levels under specific cellular conditions, it becomes possible to determine how
mammalian cells process specific Pt-DNA cross-links.
The design and construction of the parent vector, pGLuc, which contains gaussia luciferase
(GLuc) as the reporter gene under the control of CMV promoter, have been described.20 The
GLuc enzyme is an ideal reporter because it requires no post-translational modification, thus allowing rapid and responsive determination of intracellular transcription levels. Enzymatic activity can also be quantified to a high degree of sensitivity by bioluminescence assays using coelenterazine as the substrate. In addition, GLuc is naturally secreted from the cell, so determination
of expression levels can be performed by sampling cell culture media rather having to prepare
cell extracts, as with conventional luciferase assays.21,22 Transfection of transcription probes into
mammalian cells was carried out using liposomal transfecting agents. To yield meaningful data,
experiments were conducted using a cell line pair comprising a parental line and its genetic variant. This approach allowed us to study transcription profiles under conditions where specific
gene expression was either turned on or off, using the parental cell line as the control. XPF human fibroblast cells (GM08437, Coriell Institute cell depository) were used as an NER-deficient
cell model. To generate the NER-competent complement, XPF cells were stably transfected with
XPF cDNA to generate XPF-corrected XPF cells,23 designated XPFcorr.
Preparation of Transcription Probes. Globally platinated transcription probes were prepared by reacting pGLuc with varying concentrations of cisplatin in buffer.24,25 Platination levels,
6
expressed as the ratio of bound platinum per plasmid (Pt/plasmid), were determined by atomic
absorption spectroscopic measurement of the Pt content of the plasmid solution and dividing it
by the DNA concentration. A set of six plasmids with added Pt/plasmid ratios ranging from 0 to
65.0 were prepared corresponding to rb values, defined as platinum bound per nucleotide, from 0
to 8.6×10-3. Transcription experiments were performed by monitoring Gaussia-converted GLuc
levels as a function of Pt/plasmid ratio in XPF and XPFcorr cells at 8 h intervals over a period of
48 h.
Site-specifically platinated probes were prepared from pGLuc-derived plasmids by creating a
short single-stranded gap, followed by filling-in of the gap with a platinated oligonucleotide
(Figure 2). The detailed synthetic methodology was reported previously.20,26,27 pGLuc was modified with a 30-bp sequence between CMV and GLuc to prepare vectors that can accommodate
the oligonucleotide insertions. The sequence was designed such that there were two recognition
sites for Nt.BspQI nicking enzymes to create site-specific cuts on the template strand 16-nt apart,
with the intended platination site positioned between the two nicking positions. Two plasmids
designed to accommodate 1,2-d(G*pG*)-Pt and 1,3-d(G*pTpG*)-Pt adducts were prepared and
designated as pGLuc4temGG and pGLuc5temGTG, respectively (Figure 3). Detailed experimental protocols for creating single-stranded gaps and inserting the platinated oligonucleotides into
these plasmids are described in the Supporting Information. Vectors pGLuc4temGG and
pGLuc5temGTG containing cis-{Pt(NH3)2}-DNA and {Pt(dach)}-DNA cross-links were prepared, plasmids corresponding to lesions formed by cisplatin and oxaliplatin, respectively (Table
1). Unplatinated oligonucleotides were also inserted to generate control plasmids.
Transcription Profiling of Globally-Platinated Probes. The effects of platinum DNA damage on transcription were investigated using pGLuc plasmids that were globally platinated with
7
with cisplatin. In previous reports, we demonstrated using other reporter constructs that Pt-DNA
cross-links formed by cisplatin are strong inhibitors of transcription.8,25 However, in those assays, transcription levels were assayed at a single time point because cell lysis was required to
release the reporter enzyme for analysis. With the use of GLuc as the reporter enzyme, analyses
could be performed at regular time intervals on a given sample because the enzyme is extruded
into the cell media. Synchronized XPF and XPFcorr cells were transfected with the same set of
platinated probes and the cell media were collected at 8 h intervals. The cell media samples thus
obtained were analyzed for GLuc activity and normalized against unplatinated controls. Since
total cell media were collected and replaced with fresh media, total transcription levels within the
8 h interval could be quantitated.
Transcription profiles of globally platinated probes in XPF and XPFcorr cells were obtained
by plotting normalized GLuc expression levels against platination levels (Pt/plasmid ratio) as a
function of time (Figure 4). The normalized transcription levels followed an exponential relationship with respect to platination levels for both XPF and XPFcorr cells at 8 h following transfection, the first time-point at which cell media were collected. In subsequent assays, recovery of
transcription was observed for both XPF and XPFcorr cells, with a much more pronounced change
in profile for the latter. Transcription levels from probes containing low platination levels
(Pt/plasmid <10) were restored to >90% of control values in XPFcorr cells, whereas transcription
levels in XPF cells remain suppressed. In order to compare transcription inhibition between the
two cell lines quantitatively, the data were fit to an exponential curve and a D0 value, defined as
the number of Pt lesions per plasmid required to reduce transcription levels to 37% of the control,8 was extracted from the fitted curves (Figure 5). The D0 values at 8 h after transfection for
XPF and XPFcorr cells were similar, 4.0 and 5.3 respectively, which is not surprising since both
8
XPF and XPFcorr have the same genotypic origin and would have processed the probes in a similar manner. Upon further sampling, however, it was evident that recovery of transcription in
XPFcorr was much more dramatic than in the XPF cells. At the later time points, the transcription
profile could no longer be adequately modeled using an exponential curve and the D0 values obtained represented lower limits. The D0 values in XPF and XPFcorr are listed in Table 2. At 48 h,
D0 values for XPF and XPFcorr cells were approximately 10.2 and 20.7. The difference in D0 values across the time intervals and the difference in transcription profile for the same probes in two
different cell lines indicate a role for NER in the restoration of transcription.
Transcription Inhibition by Site-Specific Pt-DNA Cross-Links. In order to investigate
how specific Pt-DNA adducts inhibit transcription, we investigated site-specifically platinated
probes containing the major DNA adducts of cisplatin and oxaliplatin, namely, 1,2-d(G*pG*)-Pt
and 1,3-d(G*pTpG*)-Pt cross-links. Instead of introducing platinated lesions randomly, a single
platinated DNA lesion was inserted on the template strand between the promoter and reporter
gene sequences to create a block to RNA pol II prior to transcribing the GLuc reporter gene. The
lesion was positioned about 50 bp downstream from the transcription start site, so that the transcription complex would encounter the lesion during mRNA elongation. In keeping with experimental data in vitro, we anticipated that transcription would stall at the lesion site resulting in
a decrease in GLuc expression levels versus unplatinated controls. We also expected GLuc expression levels of the platinated probes to be lower in XPF as compared to XPFcorr cells, since
XPF cells are incapable of removing the Pt-DNA blockage. Transfections were carried out and
the transcription levels determined after 24 h, corresponding to the approximate time for a complete mammalian cell cycle. In XPF cells, transcription levels of probes containing 1,2d(G*pG*)-PtA2 and 1,2-d(G*pG*)-Pt(dach) were 50.0±7.9 % and 35.0±3.4 % of unplatinated
9
controls respectively, whereas in XPFcorr, the corresponding levels were 96.6±12.7 % and
63.4±4.5 %. For 1,3-d(G*pTpG*)-PtA2 and 1,3-d(G*pTpG*)-Pt(dach), the levels were 37.7±6.7
% and 23.7±3.2 % respectively for XPF cells and 73.2±9.2 % and 51.3±10.2 % respectively for
XPFcorr cells (Figure 6). The higher transcription levels in XPFcorr compared with XPF cells indicate that the NER-competent cells are better able to overcome the Pt-DNA transcriptional block.
DISCUSSION
Pt-DNA Cross-Links Inhibit RNA Transcription in Live Cells by Blocking RNA Elongation. Pt-DNA cross-links have previously been shown to be good inhibitors of transcription in
vitro using both cell-based assays and reconstituted systems,8,9,28-30 an observation reproduced by
the present live cell luciferase assays. There are several hypotheses about the mechanism of transcription inhibition by DNA adducts of platinum anticancer drugs.7 One hypothesis supported by
compelling structural and biochemical evidence is that Pt-DNA cross-links present a physical
impediment, or roadblock, to transcription during RNA elongation and result in stalling of the
RNA pol II enzyme at the lesion site.9,28,31 Since transcription from the globally platinated
pGLuc probes depends on accessibility of the promoter and reporter gene sequences to RNA pol
II, we expected a higher probably for the enzyme of encountering Pt lesions on probes with
higher platination levels. The data obtained indicate a strong dependence of transcription levels
on Pt/plasmid ratios in both cell lines, consistent with this hypothesis (Figure 5). To investigate
whether a single platinated DNA lesion would block RNA elongation, we engineered a Pt-DNA
cross-link site-specifically into the template strand of pGLuc between the promoter and the reporter gene sequence as a block to transcription. The cross-link was deliberately positioned 50 bp
downstream of the transcription start site so that the pol II enzyme would encounter the lesion
10
during the elongation, rather than the initiation, phase.32,33 The positioning of the cross-link upstream of the reporter gene was primarily to facilitate vector preparation, since insertion of the
artificial 30-bp platination sequence into the reporter gene sequence could adversely affect the
viability of the luciferase enzyme.
Our results reveal significant transcription inhibition using the site-specifically platinated
probes in NER-deficient XPF cells 24 h after transfection in comparison to unplatinated controls.
Thus, for the first time, we are able to demonstrate the ability of a defined adduct to block transcription in a live cell. To the extent that this activity reflects the sensitivity of cancer cells of
different lineage to platinum-DNA lesions derived from various members of the cisplatin family
of drugs, it offers the potential to personalize treatment options for future medical applications.
The results are consistent with our previous report employing run-off assays performed on
immobilized site-specifically platinated DNA templates in vitro using HeLa nuclear extracts,
which revealed that transcription is inhibited at the site of the platinated lesion.9 In addition, the
GLuc expression data indicate lower levels of transcription from the 1,2-d(G*pG*)-Pt(dach) and
1,3-d(G*pTpG*)-Pt(dach) probes by comparison to 1,2-d(G*pG*)-PtA2 and 1,3-d(G*pTpG*)PtA2, respectively, suggesting that Pt(dach)-DNA lesions are stronger transcription blocks than
PtA2-DNA adducts. This result supports the roadblock hypothesis because the bulky dach ligand
presents a greater steric hindrance than the two ammine ligands. A similar observation was made
using in vitro transcription on immobilized DNA templates containing the same Pt-DNA crosslinks by T7 RNA pol enzyme using “transcription walking.” This experiment revealed that the
enzyme completely arrests at the first G residue of the 1,2-d(G*pG*)-Pt(dach) and 1,3d(G*pTpG*)-Pt(dach) lesions on the DNA template but is able to bypass the residue of 1,2d(G*pG*)-PtA2 and 1,3-d(G*pTpG*)-PtA2 lesions, indicating more effective inhibition.34 An
11
important distinction between the present results, obtained in live cells, compared to the earlier in
vitro assay is the inability of the site-specific Pt adducts to form an absolute block to transcription in the XPF-corrected cells. Extension of such studies to cells of different genetic backgrounds and tissue type will be quite valuable for assessing the clinical potential of existing and
new members of the platinum anticancer drug family.
In summary, the data from the present luciferase live cell assay using the site-specifically
platinated probes support the hypothesis that platinated DNA adducts inhibit transcription in living cells by impeding RNA elongation. The results distinguish between non-leaving group
ligands on platinum, ammine vs. R,R-1,2-dach, and between 1,2- vs. 1,3-cross-links.
Nucleotide Excision Repair of Platinated DNA Adducts Restores Transcription. The
presence of DNA lesions caused by mutagens of endogenous or exogenous origin can have deleterious effects on cells. To defend against these insults, the cell has evolved NER as a versatile
and sophisticated mechanism to remove the damage.35 Two sub-pathways, global genomic NER
(GG-NER) and transcription-coupled NER (TC-NER), which differ in their mode of damagerecognition, can be evoked. In GG-NER, DNA damage is recognized by the XPC-hHR32B-Cen2
heterotrimeric complex, whereas in TC-NER the polymerase complex stalled at the site of a lesion is the trigger for repair.36-38 Repair by cisplatin damage is performed predominantly via the
TC-NER pathway because cells with defective TC-NER specific genes, e.g. Cockayne Syndrome
B (CS-B), are more sensitive to cisplatin treatment than cells lacking GG-NER specific genes,
e.g. XPC.37,39 In addition, with the use of plasmids globally platinated with cisplatin in HeLa nuclear extracts, it was demonstrated that the arrest of transcription on cisplatin-damaged DNA induced RNA pol II ubiquitylation.40 Since this ubiquitylation was absent in TC-NER deficient CS
cells, it was speculated that ubiquitylation is a flag to promote recruitment of NER machinery to
12
the stalled polymerase, thereby establishing a role for TC-NER in the processing of Pt-DNA
cross-links.41
Since the removal of platinated lesions should lead to the resumption of transcription, we decided to investigate changes in the transcription profile of the globally platinated probes as a
function of time. The XPF cell line is an excellent NER-deficient model because its genetic mutation leads to a defective XPF that is unable to stably heterodimerize with ERCC1 to form the
XPF/ERCC1 endonuclease complex.42 The XPF/ERCC1 endonuclease is a critical NER component responsible for creating an incision 5’ to the site of DNA damage and, therefore, XPF cells
are incapable of removing a platinated lesion from the damage site.43 It was therefore not surprising that transcription of globally platinated probes in XPF cells remained suppressed over time,
with a slight recovery in transcription only after 48 h (Figure 4). In contrast, transcription on the
same set of probes in NER-competent XPFcorr cells changed dramatically, with full recovery of
transcription in the minimally platinated probes within 48 h of transfection. To further investigate the cellular response to a platinated lesion positioned on the path of the RNA elongation
complex, transcription levels from site-specifically platinated plasmids were compared in XPF
and XPFcorr cells. After 24 h, transcription from the probes was significantly higher in XPFcorr
than XPF cells, with nearly full recovery of transcription occurring in the probe containing 1,2d(G*pG*)-PtA2 cross-link (Figure 6). The data suggest that XPFcorr cells are able to more effectively overcome a platinated lesion to restore transcription than XPF cells, consistent with earlier
findings. The data also indicate that Pt(dach) lesions are not only better inhibitors of transcription
but are less efficiently repaired than PtA2 adducts. These findings differ from that of previous
work based on excision assays using site-specifically platinated linear DNA probes with HeLa
cell-free extracts. In that report, fragments containing the platinated lesion excised from the
13
probes, which were treated with the extracts for 60 min, were resolved by polyacrylamide gel
electrophoresis and quantitated using radiography.44 No significant difference between the 1,2d(G*pG*)-Pt adducts of cisplatin and oxaliplatin were observed. In our assays, repair efficiencies
were inferred from transcription data using a NER-proficient/deficient whole cell models. Although excision assays provide a more direct measurement of repair rates, our assays carried out
in live mammalian cells provide a more realistic environment. Transfection of platinated plasmids into cell hosts other than HeLa may exhibit different degrees of transcription inhibition,
however, and we do not rule out the possibility of greater blockage of RNA pol II in live cells
than observed in the present study. The likelihood that cell type will affect the degree of transcription inhibition and the potential value of this information have already been discussed.
Repair of Pt-DNA Cross-Links by Other Mechanisms. Another subtle difference in the
data obtained with the live cell luciferase assays described in this report and run-off assays using
HeLa nuclear extracts reported previously relates to the extent of observed transcription inhibition.9,28,30 In the run-off assays, transcription was almost completely inhibited by 1,2-d(G*pG*)PtA2 and 1,3-d(G*pTpG*)-PtA2 cross-links, very low levels of run-off products detected. By
contrast, high residual transcription levels of 50.0% and 37.7% were detected for site-specifically
platinated probes in repair-deficient XPF cells using probes containing 1,2-d(G*pG*)-PtA2 and
1,3-d(G*pTpG*)-PtA2 lesions, respectively. This result indicates that cells can bypass transcription even in the absence of functioning NER machinery. In addition, transcription of globally
platinated probes recovered as a function of time in the NER-deficient XPF cells, albeit at a
slower rate compared to XPFcorr cells, suggesting removal of the Pt blockage. These observations
point to the presence of other mechanisms besides NER that can repair the Pt-DNA cross-links.
Examples of such mechanisms associated with removal of platinated DNA lesions include mis-
14
match repair and translesion synthesis, either of both of which may remove the platinum block
during the long time course of the luciferase assay.14,45-47 In vitro experiments using immobilized
platinated DNA templates and cell extracts require short exposure times to prevent degradation
of the DNA templates in the cell extracts. The present live cell luciferase experiments therefore
suggest that, whereas NER is important for the repair of platinated DNA lesions, other less
dominant repair mechanisms are also be present that are not adequately modeled by the conventional in vitro assays.
CONCLUSION
We report for the first time a system for investigating in live mammalian cells the processing
of specific Pt-DNA cross-links formed by platinum anticancer drugs using transcriptionally active expression vectors containing the programmed adducts. Our results reveal how specific PtDNA cross-links inhibit transcription in live cells, to varying degrees depending on the type of
lesion, by impeding the passage of RNA pol II, and we demonstrate that NER can effectively
remove the blockage and restore transcription. The data also indicate that there are other repair
mechanisms in operation, besides NER, that can repair Pt-DNA cross-links from the DNA template. By studying the processing of Pt-DNA cross-links within a realistic cellular environment,
our investigations represents a major step in elucidating the mechanism of action of clinically
important platinum-based drugs and offer the potential to select a platinum compound for treatment of tumors based on its ability to inhibit transcription from a globally or site-specifically
modified probe in live cells derived from the cancer tissue.
15
EXPERIMENTAL SECTION
Materials and Methods. All chemical reagents were obtained from Sigma-Aldrich unless
otherwise stated. Included are pCMV-GLuc (New England Biolabs), Lipofectamine 2000 and
OptiMEM cell media (Invitrogen), fetal bovine serum (Hyclone), coelenterazine (Nanolight
technologies), DMEM cell media and G418 sulfate (MediaTech). XPF (GM08437) cells were
obtained from the Coriell Cell Depositories (Coriell Institute, Camden, NJ, U.S.A). XPFcomplemented XPF cells (XPFcorr), originally established by stably transfecting XPF cDNA into
XPF cells, were generously donated by Dr Gan Wang (Wayne State University, Michigan). Cells
were grown in a humidified incubator at 37 °C at 5% CO2. XPF cells were maintained in DMEM
containing 4.5 g/L glucose supplemented with 10% FBS and antibiotics, and XPFcorr was maintained in DMEM containing 4.5 g/L glucose supplemented with 10% FBS, antibiotics, and 0.1
mg/mL G418 sulfate.
Quantification of Pt content on globally-platinated plasmids was carried out by graphite furnace atomic absorption spectroscopy (GF-AAS) using the Perkin-Elmer AAnalyst 300 system.
DNA concentrations were measured by UV/vis spectroscopy at 260 nm using a Varian Cary 1Espectrometer fitted with a microprobe and an extinction coefficient of 50 µg/mL/OD. The rb
value is defined here as the number of atoms of platinum complex fixed per nucleotide. For sitespecifically platinated plasmids, DNA quantification was performed using the picogreen assay
(Molecular Probes) against a plasmid DNA standard curve (0, 10, 20, 30, 40, 50 µg/mL) on the
BioTek Synergy 2 fluorescence microplate reader (λex:485 nm, λem:528 nm). Agarose gel electrophoreses (0.8-1.0% w/v) containing 0.5 µg/ml EtdBr were imaged using the BioRad Fluor-S
MultiImager.
16
Vector Construction and Preparation. A mammalian vector expressing gaussia luciferase,
pGLuc, was derived from the commercial pCMV-GLuc by PCR-mediated deletion, which was
performed in order to remove restriction sites of BspQI and the SV40 origin of replication, as
previously described.20 This procedure renders the plasmid incompetent for replication in the
host cells. pGLuc was treated sequentially with HindIII and BamHI to digest the vector between
the CMV promoter and GLuc reporter genes. It was then ligated with a synthetic insert containing sequences for 1,2-d(G*pG*)-Pt DNA lesion incorporation to yield pGLuc4temGG, as described in the Supporting Information. pGLuc5temGTG, designed to specifically incorporate a
1,3-d(G*pTpG*)-Pt adduct, was prepared from pGLuc in a similar fashion.
Preparation of Globally Platinated Plasmids. pGLuc plasmids (50 nM) were treated with
cisplatin (0.5, 1, 2, 4, 8 µM) in buffer (24 mM HEPES pH 6.0, 10 mM NaCl) for 16 h at 25 °C.
A mock platination reaction was also carried out as a control using water in place of cisplatin.
Upon completion, the reaction mixtures were dialyzed against TE/NaCl buffer (10 mM Tris·HCl
pH 8.0, 1 mM EDTA, 200 mM NaCl) for 4 h, followed by TE buffer (10 mM Tris·HCl pH 8.0, 1
mM EDTA) for 8 h with 2 exchanges of TE buffer solution. The rb values of plasmids were determined to be 0, 2.8×10-4, 6.9×10-4, 1.2×10-3, 4.3×10-3, 8.6×10-3, corresponding to 0, 2.1, 5.2,
9.2, 32.6, 65.0 Pt per plasmid.
Preparation of Site-Specifically Platinated Plasmids. Site-specifically platinated plasmids
containing a single Pt-DNA lesion positioned between the CMV promoter and Gaussia luciferase
reporter gene were prepared as described previously.26 Experimental details are provided in the
Supporting Information. Incorporation of 1,2-d(G*pG*)-Pt and 1,3-d(G*pTpG*)-Pt lesioncontaining oligonucleotides was carried out on pGLuc4temGG and pGLuc5temGTG plasmids
respectively. Plasmids containing cisplatin-type cis-{Pt(NH3)2}-DNA adducts were designated
17
with the “+is-PtA2” insertion strand suffix, and plasmids with oxaliplatin-type {Pt(dach)}-DNA
adducts as “+is-Pt(dach)”. Control plasmids, prepared by mock insertion of unplatinated oligonucleotides, were designated with an “+is” suffix. A summary of plasmids used in this study is
provided in Table 1.
Transient transfection of cells for transcription assays. Transcription assays were carried
by transient transfection of the plasmids into XPF and XPFcorr cells, synchronized at G1/S phase,
followed by measurement of the levels of luciferase reporter gene expression. Cell synchronization at G1/S was accomplished using a double thymidine block. Briefly, XPF and XPFcorr cells
were distributed on a 96-well plate at a density of 5,000 cells per well. At 24 h after cell plating,
cell media were aspirated and fresh pre-warmed media containing 2.5 mM thymidine were
added. After a 19 h incubation period, the cell media were replaced and the cells were incubated
for 12 h under thymidine-free conditions. The cell media were then aspirated, and fresh prewarmed media containing 2.5 mM thymidine were added. The cells were incubated for an additional 12 h to achieve G1/S synchronization and used for the transcription experiments.
Transient transfection of the cells was carried out using Lipofectamine 2000 as the transfection agent and performed in quadruplicate. The transfection solutions were prepared as a 5X
master mix since it was impractical to handle minute volumes of probes and transfection agent.
Briefly, reconstituted solutions of plasmid DNA (500 ng for globally platinated probes and 50 ng
for site-specifically platinated probes) in Opti-MEM (62.5 µL) and lipofectamine 2000 (1.25 µL)
in OptiMEM (62.5 µL) were prepared separately. After 5 min incubation, the two solutions were
combined and incubated for 30 min, during which time the liposome-DNA complexation took
place. The liposomal mixture was diluted with 250 µL antibiotics-free culture media (DMEM
with 10% FBS) to yield 375 µL 5X master mix transfection solution.
18
Prior to transfection, synchronized XPF and XPFcorr cells were washed twice with antibioticsfree culture media (100 µL). Transfection solutions (75 µL) were transferred to each well and the
cells were incubated for 2 h. To remove the transfection solution, cells were washed twice with
pre-warmed antibiotics-free culture media (100 µL) and returned to the incubator. This time
point was designated as the start of the experiment (t = 0). For experiments involving the site
specifically platinated probes, cell media were collected at t = 24 h. For experiments involving
globally platinated probes, cell media were collected at 8 h intervals over a 48 h period and the
media were replaced with fresh pre-warmed antibiotics-free culture media. Cell media from untransfected cells were also collected as background control. The collected cell media were stored
at 4 °C.
GLuc reporter gene assays. GLuc expression levels were quantitated by using an assay that
measures bioluminescence as a function of GLuc enzymatic activity in the presence of coelenterazine substrate. Collected cell media (10 µL) were transferred to 96-well white opaque assay
plates. A luminescence plate reader (BioTek), fitted with reagent auto-injectors, was employed to
assay GLuc activity of each individual well. Briefly, a GLuc assay solution (25 µL, 10 mM
Tris·HCl pH 7.8, 1 mM EDTA, 600 mM NaCl, 10 mM coelenterazine) was injected into each
well. The plate was allowed to shake for 5 s, after which a luminescence reading was taken. The
readings were corrected against background readings taken from untransfected cells.
ACKNOWLEDGEMENT
The project described was supported by grant CA032134 from the National Cancer Institute.
The authors thank Prof. Gan Wang (Wayne State University) for a generous donation of XPF-
19
complemented XPF cells. WHA thanks the National University of Singapore for an Overseas
Postdoctoral Fellowship.
Supporting Information Available: Preparation and characterization of site-specifically platinated DNA plasmids. This material is available free of charge via the Internet at
http://pubs.acs.org.
20
REFERENCES
(1)
Kelland, L. Nat. Rev. Cancer 2007, 7, 573-584.
(2)
Horwich, A.; Shipley, J.; Huddart, R. Lancet 2006, 367, 754-765.
(3)
Keys, H. M.; Bundy, B. N.; Stehman, F. B.; Muderspach, L. I.; Chafe, W. E.; Suggs, C.
L., 3rd; Walker, J. L.; Gersell, D. N. Engl. J. Med. 1999, 340, 1154-1161.
(4)
Jamieson, E. R.; Lippard, S. J. Chem. Rev. 1999, 99, 2467-2498.
(5)
Wang, D.; Lippard, S. J. Nat. Rev. Drug Discover. 2005, 4, 307-320.
(6)
Jung, Y.; Lippard, S. J. Chem. Rev. 2007, 107, 1387-1407.
(7)
Todd, R. C.; Lippard, S. J. Metallomics 2009, 1, 280-291.
(8)
Mello, J. A.; Lippard, S. J.; Essigmann, J. M. Biochemistry 1995, 34, 14783-14791.
(9)
Jung, Y.; Lippard, S. J. J. Biol. Chem. 2006, 281, 1361-1370.
(10)
Husain, I.; Chaney, S. G.; Sancar, A. J. Bacteriol. 1985, 163, 817-23.
(11)
Zamble, D. B.; Mu, D.; Reardon, J. T.; Sancar, A.; Lippard, S. J. Biochemistry 1996, 35,
10004-10013.
(12)
Plooy, A. C. M.; van Dijk, M.; Berends, F.; Lohman, P. H. M. Cancer Res. 1985, 45,
4178-4184.
(13)
Moggs, J. G.; Yarema, K. J.; Essigmann, J. M.; Wood, R. D. J. Biol. Chem. 1996, 271,
7177-7186.
(14)
Moggs, J. G.; Szymkowski, D. E.; Yamada, M.; Karran, P.; Wood, R. D. Nucleic Acids
Res. 1997, 25, 480-490.
(15)
Cohen, S. M.; Jamieson, E. R.; Lippard, S. J. Biochemistry 2000, 39, 8259-8265.
(16)
Wei, M.; Cohen, S. M.; Silverman, A. P.; Lippard, S. J. J. Biol. Chem. 2001, 276, 3877438780.
21
(17)
Jung, Y.; Mikata, Y.; Lippard, S. J. J. Biol. Chem. 2001, 276, 43589-43596.
(18)
Shivji, M. K. K.; Moggs, J. G.; Kuraoka, I.; Wood, R. D. Methods Mol. Biol. 2006, 314,
435-456.
(19)
Malina, J.; Novakova, O.; Vojtiskova, M.; Natile, G.; Brabec, V. Biophys. J. 2007, 93,
3950-3962.
(20)
Ang, W. H.; Lippard, S. J. Chem. Commun. 2009, 5820-5822.
(21)
Tannous, B. A.; Kim, D.-E.; Fernandez, J. L.; Weissleder, R.; Breakefield, X. O. Mol.
Ther. 2005, 11, 435-443.
(22)
Michelini, E.; Cevenini, L.; Mezzanotte, L.; Ablamsky, D.; Southworth, T.; Branchinic,
B. R.; Roda, A. Photochem. Photobiol. Sci. 2008, 7, 212-217.
(23)
Chen, Z.; Xu, X. S.; Harrison, J.; Wang, G. Biochem. J. 2004, 379, 71-78.
(24)
Todd, R. C.; Lovejoy, K. S.; Lippard, S. J. J. Am. Chem. Soc. 2007, 129, 6370-6371.
(25)
Lovejoy, K. S.; Todd, R. C.; Zhang, S.; McCormick, M. S.; D'Aquino, J. A.; Reardon, J.
T.; Sancar, A.; Giacomini, K. M.; Lippard, S. J. Proc. Natl. Acad. Sci. U. S. A. 2008, 105,
8902-8907.
(26)
Ang, W. H.; Brown, W. W.; Lippard, S. J. Bioconjug. Chem. 2009, 20, 1058-1063.
(27)
Wang, H.; Hays, J. B. Mol. Biotechnol. 2001, 19, 133-140.
(28)
Tremeau-Bravard, A.; Riedl, T.; Egly, J.-M.; Dahmus, M. E. J. Biol. Chem. 2004, 279,
7751-7759.
(29)
Cullinane, C.; Mazur, S. J.; Essigmann, J. M.; Phillips, D. R.; Bohr, V. A. Biochemistry
1999, 38, 6204-6212.
(30)
Tornaletti, S.; Patrick, S. M.; Turchi, J. J.; Hanawalt, P. C. J. Biol. Chem. 2003, 278,
35791-35797.
22
(31)
Damsma, G. E.; Alt, A.; Brueckner, F.; Carell, T.; Cramer, P. Nat. Struct. Mol. Biol.
2007, 14, 1127-1133.
(32)
Dahmus, M. E. J. Biol. Chem. 1996, 271, 19009-19012.
(33)
Bartholomew, B.; Dahmus, M. E.; Meares, C. F. J. Biol. Chem. 1986, 261, 14226-14231.
(34)
Jung, Y.; Lippard, S. J. J. Biol. Chem. 2003, 278, 52084-52092.
(35)
de Laat, W. L.; Jaspers, N. G. J.; Hoeijmakers, J. H. J. Genes Dev. 1999, 13, 768-785.
(36)
Iyer, N.; Reagan, M. S.; Wu, K.-J.; Canagarajah, B.; Friedberg, E. C. Biochemistry 1996,
35, 2157-2167.
(37)
van Gool, A. J.; Citterio, E.; Rademakers, S.; van Os, R.; Vermeulen, W.; Constantinou,
A.; Egly, J.-M.; Bootsma, D.; Hoeijmakers, J. H. J. EMBO J. 1997, 16, 5955-5965.
(38)
Hoogstraten, D.; Bergink, S.; Ng, J. M. Y.; Verbiest, V. H.; Luijsterburg, M. S.; Geverts,
B.; Raams, A.; Dinant, C.; Hoeijmakers, J. H. J.; Vermeulen, W.; Houtsmuller, A. B. J.
Cell Sci. 2008, 121, 2850-2859.
(39)
Furuta, T.; Ueda, T.; Aune, G.; Sarasin, A.; Kraemer, K. H.; Pommier, Y. Cancer Res.
2002, 62, 4899-4902.
(40)
Lee, K.-B.; Wang, D.; Lippard, S. J.; Sharp, P. A. Proc. Natl. Acad. Sci. U. S. A. 2002,
99, 4239-4244.
(41)
Bregman, D. B.; Halaban, R.; van Gool, A. J.; Henning, K. A.; Friedberg, E. C.; Warren,
S. L. Proc. Natl. Acad. Sci. U. S. A. 1996, 93, 11586-11590.
(42)
Matsumura, Y.; Nishigori, C.; Yagi, T.; Imamura, S.; Takebe, H. Hum. Mol. Genet. 1998,
7, 969-974.
(43)
Ferry, K. V.; Hamilton, T. C.; Johnson, S. W. Biochem. Pharmacol. 2000, 60, 13051313.
23
(44)
Reardon, J. T.; Vaisman, A.; Chaney, S. G.; Sancar, A. Cancer Res. 1999, 59, 3968-3971.
(45)
Yamada, M.; O'Regan, E.; Brown, R.; Karran, P. Nucleic Acids Res. 1997, 25, 491-495.
(46)
Hoffmann, J.-S.; Pillaire, M.-J.; Lesca, C.; Burnouf, D.; Fuchs, R. P. P.; Defais, M.; Villani, G. Proc. Natl. Acad. Sci. U. S. A. 1996, 93, 13766-13769.
(47)
Hoffmann, J.-S.; Locker, D.; Villani, G.; Leng, M. J. Mol. Biol. 1997, 270, 539-43.
24
FIGURES
Figure 1. Chemical structures of platinum anticancer drugs.
Figure 2. Scheme for preparing site-specifically platinated probes.
25
Figure 3. Site-specifically platinated probes containing Pt-DNA cross-links of cisplatin and oxaliplatin; platination sites are highlighted in bold.
Figure 4. Transcription profile of globally platinated probes in XPF (left) and XPFcorr (right)
cells.
26
Figure 5. Comparison of transcription profiles of globally platinated probes in XPF and XPFcorr
cells 8 h (left) or 48 h (right) after transfection.
Figure 6. Transcription of site-specifically platinated probes containing 1,2-d(G*pG*)-Pt
(pGLuc4temGG, left) and 1,3-d(G*pTpG*)-Pt (pGLuc5temGTG, right) cross-links 24 h after
transfection.
27
TABLES
Table 1. Designation of Site-Specifically Platinated Plasmids
Type of lesion
1,2-d(G*pG*)-Pt
1,3-d(G*pTpG*)-Pt
Plasmid designation
pGluc4temGG
pGluc5temGTG
Unplatinated control
+is
+is
Cisplatin-adduct cis-{Pt(NH3)2}
+is-PtA2
+is-PtA2
Oxaliplatin-adduct {Pt(dach)}
+is-Pt(dach)
+is-Pt(dach)
Table 2. D0 Values of Globally Platinated Probes Assayed at Different Time Intervals After
Transfectiona
a
Time after
XPF
XPFcorr
transfection
(Pt/plasmid)
(Pt/plasmid)
8h
4.0
5.4
16 h
5.3
7.9
24 h
7.6
11.5
32 h
8.7
15.4
40 h
10.0
18.7
48 h
10.2
20.7
D0 value is defined as the number of Pt lesions per plasmid required to reduce transcription lev-
els to 37% of the control
28
TOC GRAPHIC